The GeekNarrator

The GeekNarrator

By Kaivalya ApteTechnology
Download on the App Store

The GeekNarrator episodes

  • Batch vs Realtime Stream Processing - A Deep Dive with Phil Fried from Estuary

    In this video I talk to Philip Fried from Estuary about Batch vs Realtime Stream Processing.

    Philip brings a ton of experience in the world of data processing and has shared some of the best practices in implementing these systems. We dive deep into the world of data processing, covering batch and streaming systems, their challenges, tradeoffs and use cases.
    Chapters:
    00:00 Batch vs Realtime Stream Processing
    03:25 What is Batch and Reatlime processing?
    18:29 How does Batch and Realtime compare in terms of Latency and Throughput?
    27:24 Where is the cost saving coming from? Compute?Storage? or Network?
    31:38 Moving from Batch to Stream processing
    37:50 How is Idempotency implemented in Streaming systems?
    48:50 How do we approach Schema evolution in Batch and Streaming systems?
    57:16 Summary - key points to keep in mind
    Do checkout Estuary if you deal with a ton of data, and don't want to deal with the painful operations, infrastructure management, schema migrations etc and only want to focus on building highly scalable and resilient applications.
    References:
    Estuary: https://estuary.dev/
    Flow documentation: https://docs.estuary.dev
    If you like this video please hit the like button, share it with your network (whoever works with a ton of data) and subscribe to the channel.
    Feel free to watch related episodes in the playlist:    • Distributed Syste...  
    Modern Databases:    • Modern Databases  
    Software Engineering:    • Software Engineering  
    Distributed Systems:    • Distributed Systems  
    Cheers,
    The GeekNarrator

    1 hr 4 min
  • Cassandra 5.0 - ACID transactions and Vector Search

    Hey Everyone,

    In this video I talk to Patrick McFadin from DataStax. We uncovered the new features in Cassandra 5.0 and discussed how ACID transactions are achieved in the new version. This is a deep dive into the features of Cassandra, consensus protocols and how Accord is different as compared to Paxos, RAFT, Spanner and Calvin.
    Chapters:
    Cassandra 5.0 - ACID transactions and Vector Search
    00:00 Introduction
    01:45 List of features in the new Cassandra
    04:51 Who needs ACID properties?
    07:20 Why didn't Cassandra have ACID properties so far?
    10:35 Why is Accord consensus protocol well suited for Cassandra?
    16:40 Lets take a gaming example to see how Transactions work
    21:55 Whats happening behind the scenes in a Transactions?
    27:44 What happens when there are failures?
    33:41 How is upgrade to the new version going to look like?
    35:48 How is the latency impacted because of transactions?
    40:23 What was missing in lightweight transactions?
    42:24 Vector Search - What is it? How does it work?
    Previous episode on Cassandra:    • Apache Cassandra ...  
    Other playlists to watch:
    Distributed systems and Databases:    • Distributed Syste...  
    Software Engineering:    • Software Engineering  
    Distributed systems:    • Distributed Systems  
    Modern Databases:    • Modern Databases  
    Patrick's Linkedin: https://www.linkedin.com/in/patrick-m...
    Astra: astra.datastax.com
    Cassandra: https://cassandra.apache.org/_/index....
    I hope you liked this episode. If you did, please hit the like button, share it with your network and subscribe to the channel.
    Cheers,
    The GeekNarrator

    51 min
  • Designing a Modern Ledger Database with Michael Parsons (CTO and Co-Founder @Twisp)

    Hey Everyone,

    In this episode I discuss Twisp's journey with Mike Parsons, CTO and Co-Founder of Twisp. We have discussed how they ended up building a cloud native distributed ledger database on top of DynamoDB, why they chose DynamoDB? what were the challenges they faced and many highly technical and very interesting topics.
    Chapters:
    00:00 Introduction - Twisp - CTO and Co-Founder Mike Parsons
    03:01 What do we mean by a Distributed Ledger Database?
    07:13 Is it similar to Blockchain?
    08:32 Challenges faced with Relational Database setup
    19:39 Why DynamoDB was the right choice?
    24:40 What was the scale you were building for?
    26:05 Who uses a Ledger database?
    29:30 What are the basic primtives in a ledger database?
    33:55 Is every primitive Immutable? How is it stored?
    38:07 Impact of Immutability on Data storage
    41:33 How does idempotency work in Twisp?
    44:53 How did you provide ACID guarantees on top of DynamoDB?
    52:30 How do you ensure correctness?
    57:13 Interesting real world usecases where Twisp is used.
    01:02:33 Are you planning to build even higher level primitives?
    References:
    Clocks: https://cse.buffalo.edu/tech-reports/... and http://rystsov.info/2018/10/01/tso.html
    TicToc: https://people.csail.mit.edu/sanchez/...
    Hekaton Paper: https://www.microsoft.com/en-us/resea...
    Hermitage Test suite: https://github.com/ept/hermitage
    Online Async Schema Change in F1 : https://static.googleusercontent.com/...
    Designing Data intensive Applications: https://www.oreilly.com/library/view/...
    P Programming language: https://github.com/p-org/P
    Twisp: https://www.twisp.com/
    AWS Dynamo DB: https://aws.amazon.com/pm/dynamodb/?t...
    Connect with Mike:
    https://www.linkedin.com/in/parsnips/
    Follow me:
    Twitter: https://twitter.com/thegeeknarrator
    Linkedin: https://www.linkedin.com/in/kaivalyaa...
    Watch other playlists:
    Databases:    • Modern Databases  
    Distributed systems:    • Distributed Systems  
    Software Engineering:    • Software Engineering  
    Sofk skills:    • Soft Skills  
    Like, share and subscribe to the channel folks. See you in the other episode.
    Cheers,
    The GeekNarrator

    1 hr 9 min
  • Serverless Architecture with Yan Cui

    In this episode I talk to Yan Cui, who is an AWS Serverless Hero, all about Serverless technologies.

    Chapters:
    00:00 Serverless Architecture with Yan Cui
    01:58 What do we mean by Serverless Architecture?
    05:42 What is the core problem Serverless solves?
    11:06 Do we need to think differently to be able to use Serverless?
    15:27 What is the difference between serverless and managed services?
    19:17 Is Vendor Lock-in really a problem?
    27:42 Multicloud - Is it really worth it?
    33:46 Is ColdStart a real problem? What kind of apps get impacted?
    43:25 Monitoring serverless applications
    48:22 Usecases when serverless may not be the best solution
    54:27 Future of serverless
    57:31 How should a developer learn about serverless?
    I hope you enjoy the discussion and learn from it. Please hit the like button, share it with your network and also subscribe to the channel.
    References:
    Yan Cui - https://theburningmonk.com
    Courses - https://productionreadyserverless.com/
    Corey Quinn on MultiCloud -    • Corey Quinn: The ...  
    Linkedin Yan - https://www.linkedin.com/in/theburnin...
    Twitter Yan - https://twitter.com/theburningmonk
    Other playlists to watch:
    Distributed Systems and Databases -    • Distributed Syste...  
    Software Engineering -    • Software Engineering  
    Distributed systems practices -    • Distributed Systems  
    Cheers,
    The GeekNarrator

    1 hr
  • Engineering, Productivity and Learning with Dr. Venkat Subramaniam

    It was an honour to host Dr. Venkat Subramaniam on the podcast.

    Dr. Venkat Subramaniam is an award-winning author, founder of Agile Developer, Inc., and an instructional professor at the University of Houston. He has mentored tens of thousands of software developers in the US, Canada, Europe, and Asia, and is a regularly-invited speaker at several international conferences. Venkat helps his clients effectively apply and succeed with agile practices on their software projects.
    In this episode, Dr. Venkat has shared his experience on becoming productive, managing time effectively, learning and Software Engineering. There is so much to learn from the way Venkat approaches learning and keeping up with the pace. No wonder, how he is able to do so many different things.
    Chapters:
    00:00:00 Introduction into Venkat's Journey
    00:04:23 How is Venkat able to do so many things?
    00:15:23 Saying NO and prioritisation
    00:44:42 Keeping up with the pace and learn
    00:53:55 Engineering practices every engineer must follow
    01:14:40 Ending notes, resources etc..
    This episode is full of stories, experiences and learnings from Dr. Venkat's vast experience. The analogies he has shared are amazing.
    I hope you like this discussion and also learn a ton of things that matter.
    Venkat Subramaniam: https://twitter.com/venkat_s
    Agile Developer: https://agiledeveloper.com/
    If you like this discussion, please like, share, subscribe and press the bell icon. Keep supporting and more amazing discussions coming your way.
    Cheers,
    The GeekNarrator

    1 hr 22 min
  • Functional Programming on the JVM with Daniel from RockTheJVM

    Hey Everyone,

    In this video I was joined by Daniel Ciocîrlan from https://rockthejvm.com and we discussed Functional Programming concepts in great detail.
    Chapters:
    00:00 Introduction
    03:11 What is Functional Programming? How can I start learning?
    11:40 Object Oriented Programming and Functional Programming
    16:08 How should a beginner approach learning Functional Programming?
    21:58 Handling Side Effects (Pure functions)
    28:04 Immutability and Recursion
    35:59 Concurrency Models (Futures, Co-routines, Actors)
    42:47 Project Loom
    46:42 Scala vs Kotlin
    49:01 Current Limitations and Future
    RockTheJVM : https://rockthejvm.com
    I hope you like the episode, please like, share and subscribe for more videos.
    Cheers,
    The GeekNarrator

    56 min
  • Time Management for Software Engineers

    Hello Everyone,

    Time Management is a critical skill for Software Engineers and most of us struggle with it everyday. So I and Arpit decided to discuss and share our learnings and experience on time management for Software Engineers.
    We have discussed about:
    - How to balance Context Switching?
    - How to say NO?
    - How and when to ask for help?
    - Prioritisation of tasks
    - Visibility of your work
    - Documentation and helping others
    - Getting better at Time Management
    I hope this casual discussion will help you figure out gaps in your approach and help you get better at managing your time well :)
    If you like the episode, please like, share and subscribe to the channel. Also don't forget to press the bell icon to get notified on any new video added to the series.
    Arpit Bhayani : https://www.linkedin.com/in/arpitbhay... and https://arpitbhayani.me/


    #softwareengineering #timemanagement

    Cheers,
    The GeekNarrator

    34 min
  • Building Robust Data Platforms with Francesco Tisiot

    Hey Everyone,


    In this episode I was joined by Francesco Tisiot who is a Developer Advocate @Aiven. He has shared some interesting insights into how you can build robust data platforms. The approach we have discussed is also described here: https://aiven.io/blog/a-soft-methodology-to-define-robust-data-platforms


    References:   

    SOFT : https://aiven.io/blog/a-soft-methodology-to-define-robust-data-platforms

    Aiven: https://aiven.io/Francesco 

    Tisiot https://twitter.com/FTisiot and https://www.linkedin.com/in/francescotisiot/

    I hope you like the episode. Please subscribe to the channel and share with your network.   


    Cheers, The GeekNarrator

    1 hr 14 min
  • Understanding ksqlDB with Matthias J. Sax

    Hey Everyone,  

    In this episode I and Matthias talk about KsqlDb. We have covered the topic in great depth talking about its history, architecture, different concepts, use cases, limitations, comparison to Kafka Streams and so on.

    References: 

    ksqlDB - https://ksqldb.io/

    exactly once semantics podcast: https://youtu.be/twgbAL_EaQw

    Matthias Sax: https://twitter.com/MatthiasJSax and https://www.linkedin.com/in/mjsax/ 

    Cheers, The GeekNarrator

    1 hr 2 min
  • Building Geo Distributed Applications with Denis Magda

    Hey Everyone, 

    In this episode I have Denis Magda from YugabyteDb with me.

    We have explained how to build Geo-Distributed Applications in depth.  

    I am sure this will help you make your applications more scalable, highly resilient, compliant and highly performant. If you liked the video please subscribe to the channel and give me a like.  


    Further reading: 

    https://www.yugabyte.com/blog/what-is... 

    https://cloud.google.com/load-balancing

    AWS outage: https://aws.amazon.com/message/41926/

     Connect with Denis: https://www.linkedin.com/in/dmagda/ and https://twitter.com/denismagda 

    Connect with me: https://www.linkedin.com/in/kaivalya-... 

    Cheers, The GeekNarrator

    1 hr 14 min

About The GeekNarrator

From the publisher's feed

The GeekNarrator podcast is a show hosted by Kaivalya Apte who is a Software Engineer and loves to talk about Technology, Technical Interviews, Self Improvement, Best Practices and Hustle.